Stat Project

Source file ⇒ rosa_graphs.Rmd

UC_arrests_noncampus = read.csv("NorCal SoCal/Public_UC_Arrests_Noncampus.csv")
UC_arrests_campus = read.csv("NorCal SoCal/Public_UC_Arrests_On_Campus.csv")
UC_arrests_housing = read.csv("NorCal SoCal/Public_UC_Arrests_On_campus_Student_Housing_Facilities.csv")

UC_criminal_noncampus = read.csv("NorCal SoCal/Public_UC_Criminal_Offenses_Noncampus.csv")
UC_criminal_campus = read.csv("NorCal SoCal/Public_UC_Criminal_Offenses_On_campus.csv")
UC_criminal_housing = read.csv("NorCal SoCal/Public_UC_Criminal_Offenses_On_campus_Student_Housing_Facilities.csv")

UC_disciplinary_noncampus = read.csv("NorCal SoCal/Public_UC_Disciplinary_Actions_Noncampus.csv")
UC_disciplinary_campus = read.csv("NorCal SoCal/Public_UC_Disciplinary_Actions_On_campus.csv")
UC_disciplinary_housing = read.csv("NorCal SoCal/Public_UC_Disciplinary_Actions_Student_Housing_Facilities.csv")

NUC_arrests_noncampus = read.csv("NorCal SoCal/Public_NUC_Arrests_Noncampus.csv")
NUC_arrests_campus = read.csv("NorCal SoCal/Public_NUC_Arrests_On_campus.csv")
NUC_arrests_housing = read.csv("NorCal SoCal/Public_NUC_Arrests_On_campus_Student_Housing_Facilities.csv")

NUC_criminal_noncampus = read.csv("NorCal SoCal/Public_NUC_Criminal_Offenses_Noncampus.csv")
NUC_criminal_campus = read.csv("NorCal SoCal/Public_NUC_Criminal_Offenses_On_campus.csv")
NUC_criminal_housing = read.csv("NorCal SoCal/Public_NUC_Criminal_Offenses_On_campus_Student_Housing_Facilities.csv")

NUC_disciplinary_noncampus = read.csv("NorCal SoCal/Public_NUC_Disciplinary_Actions_Noncampus.csv")
NUC_disciplinary_campus = read.csv("NorCal SoCal/Public_NUC_Disciplinary_Actions_On_campus.csv")
NUC_disciplinary_housing = read.csv("NorCal SoCal/Public_NUC_Disciplinary_Actions_Student_Housing_Facilities.csv")

Arrests <- c("Weapons", "Drugs", "Liquor")
ArrestsPerStudent <- paste(Arrests, "PerStudent", sep="")

Crimes <- c("ForcibleSexOffenses", "Assault", "Burglary", "CarTheft", "Arson", "Total")
CrimesPerStudent <- paste(Crimes, "PerStudent", sep="")

VAWA <- c("DomesticViolence", "DatingViolence", "Stalking", "Rape", "Fondling", "StatutoryRape")
VAWAPerStudent <- paste(VAWA, "PerStudent", sep="")

filterArrests <- function(data) {
  return(data %>% filter(!grepl("Hastings", data$Institution.name)) %>%
                                          filter(!is.na(Institution.Size)) %>% 
                                          mutate(Weapons = as.numeric(Illegal.weapons.possession),
                                                 Drugs = as.numeric(Drug.law.violations),
                                                 Liquor = as.numeric(Liquor.law.violations)) %>%
                                          mutate(WeaponsPerStudent = Weapons/Institution.Size, 
                                                 DrugsPerStudent = Drugs/Institution.Size, 
                                                 LiquorPerStudent = Liquor/Institution.Size) %>%
                                          mutate(Total = Weapons + Drugs + Liquor, TotalPerStudent = Total/Institution.Size) %>%
                                          select(Survey.year, Institution.name, one_of(Arrests), one_of(ArrestsPerStudent)))
}

filterCriminal <- function(data){
   data %>% filter(!grepl("Hastings", data$Institution.name)) %>%
                                            filter(!is.na(Institution.Size)) %>% 
                                            mutate(ForcibleSexOffenses = as.numeric(Sex.offenses...Forcible),
                                                   Assault = as.numeric(Aggravated.assault),
                                                   CarTheft = as.numeric(Motor.vehicle.theft)) %>%
                                            mutate(ForcibleSexOffensesPerStudent = ForcibleSexOffenses/Institution.Size,
                                                   AssaultPerStudent = Assault/Institution.Size,
                                                   BurglaryPerStudent = Burglary/Institution.Size,
                                                   CarTheftPerStudent = CarTheft/Institution.Size,
                                                   ArsonPerStudent = Arson/Institution.Size) %>%
                                            mutate(Total = ForcibleSexOffenses + Assault + Burglary + CarTheft + Arson,
                                                   TotalPerStudent = Total/Institution.Size) %>%
                                            select(Survey.year, Institution.name, one_of(Crimes), one_of(CrimesPerStudent))
}

filterVAWA <- function(data) {
  data %>% filter(!grepl("Hastings", data$Institution.name)) %>%
                                            filter(!is.na(Institution.Size)) %>% 
                                            mutate(DomesticViolence = as.numeric(Domestic.violence),
                                                   DatingViolence = as.numeric(Dating.violence),
                                                   StatutoryRape = as.numeric(Statutory.Rape)) %>%
                                            mutate(DomesticViolencePerStudent = DomesticViolence/Institution.Size,
                                                   DatingViolencePerStudent = DatingViolence/Institution.Size,
                                                   StalkingPerStudent = Stalking/Institution.Size,
                                                   RapePerStudent = Rape/Institution.Size,
                                                   FondlingPerStudent = Fondling/Institution.Size,
                                                   StatutoryRapePerStudent = StatutoryRape/Institution.Size) %>%
                                            mutate(Total = DomesticViolence + DatingViolence + Stalking + Rape + Fondling + StatutoryRape) %>%
                                            select(Survey.year, Institution.name, one_of(VAWA), one_of(VAWAPerStudent))
}

sumSexOffenses <- function(data) {
  return(data %>% mutate(Sex.offenses...Forcible = ifelse(Survey.year > 2013, Rape + Fondling, Sex.offenses...Forcible)) %>%
          mutate(Sex.offenses...Non.forcible = ifelse(Survey.year > 2013, Incest + Statutory.rape, Sex.offenses...Non.forcible)))
}

updatedNames <- function(data, UC=TRUE) {
  if (UC) {
    return(gsub(".*-", "", data$Institution.name))
  } else {
    data$Institution.name <- gsub("California State University", "CSU", data$Institution.name)
    data$Institution.name <- gsub(".*Polytechnic.*-", "CalPoly-", data$Institution.name)
    return(data$Institution.name)
  }
  
}

getBerkeley <- function(data) {
  return(data %>% filter(grepl("Berkeley", Institution.name)) %>% ungroup())
}

sumNonBerkeley <- function(data, UC=TRUE) {
  data = data %>% filter(!grepl("Berkeley", Institution.name)) %>% 
                                                group_by(Survey.year) %>% 
                                                summarise_each(funs(mean))

  if (UC) {
     return(data %>% mutate(Institution.name = "Other UCs"))
  } else {
    return(data %>% mutate(Institution.name = "Non-UCs"))
  }
}

massPlot <- function(data, cols) {
  for(col in cols) {
    print(data %>% ggplot(aes(x = Survey.year, col=Institution.name)) + geom_line(aes_string(y=col)))
  }
}

groupCriminal <- function(data1, data2, data3, UC=TRUE){
  data <- rbind(data1, data2, data3)
  data$Institution.name = updatedNames(data, UC)
  return(data %>% sumSexOffenses() %>%
                                filterCriminal() %>%
                                group_by(Survey.year, Institution.name) %>%
                                summarise_all(.funs=sum) %>%
                                ungroup(Survey.year, Institution.name))
}

groupArrests <- function(data1, data2, data3, UC=TRUE){
  data <- rbind(data1, data2, data3)
  data$Institution.name = updatedNames(data, UC)
  return(data %>% filterArrests() %>%
                                group_by(Survey.year, Institution.name) %>%
                                summarise_all(.funs=sum) %>%
                                ungroup(Survey.year, Institution.name))
}

UC_criminal <- groupCriminal(UC_criminal_campus, UC_criminal_housing, UC_criminal_noncampus, UC=TRUE)

UC_arrests <- groupArrests(UC_arrests_campus, UC_arrests_housing, UC_arrests_noncampus, UC=TRUE)

UC_disciplinary <- groupArrests(UC_disciplinary_campus, UC_disciplinary_housing, UC_disciplinary_noncampus, UC=TRUE)

NUC_criminal <- groupCriminal(NUC_criminal_campus, NUC_criminal_housing, NUC_criminal_noncampus, UC=FALSE)

NUC_arrests <- groupArrests(NUC_arrests_campus, NUC_arrests_housing, NUC_arrests_noncampus, UC=FALSE)

NUC_disciplinary <- groupArrests(NUC_disciplinary_campus, NUC_disciplinary_housing, NUC_disciplinary_noncampus, UC=FALSE)

Berkeley_arrests = getBerkeley(UC_arrests)

Other_UC_arrests = sumNonBerkeley(UC_arrests)

summedUC_arrests = rbind(Berkeley_arrests, Other_UC_arrests)

massPlot(UC_arrests, Arrests)

massPlot(UC_arrests, ArrestsPerStudent)

massPlot(summedUC_arrests, Arrests)

massPlot(summedUC_arrests, ArrestsPerStudent)

Berkeley_criminal = getBerkeley(UC_criminal)

Other_UC_criminal = sumNonBerkeley(UC_criminal)

summedUC_criminal = rbind(Berkeley_criminal, Other_UC_criminal)

massPlot(UC_criminal, Crimes)

massPlot(UC_criminal, CrimesPerStudent)

massPlot(summedUC_criminal, Crimes)

massPlot(summedUC_criminal, CrimesPerStudent)

Berkeley_disciplinary = getBerkeley(UC_disciplinary)

Other_UC_disciplinary = sumNonBerkeley(UC_disciplinary)

summedUC_disciplinary = rbind(Berkeley_disciplinary, Other_UC_disciplinary)

massPlot(UC_disciplinary, Arrests)

massPlot(UC_disciplinary, ArrestsPerStudent)

massPlot(summedUC_disciplinary, Arrests)

massPlot(summedUC_disciplinary, ArrestsPerStudent)

Disciplinary Actions vs. Arrests (Students vs. Community)

compareDiscplinaryArrests <- function(disciplinary, arrests) {
  combined <- merge(disciplinary, arrests, by=c("Survey.year", "Institution.name"))
  UC_criminal$Institution.name = gsub(".*-", "", UC_criminal$Institution.name)
  names(combined) <- gsub("y$","Arrests", names(combined))
  names(combined) <- gsub("x$","Disciplinary", names(combined))
  for(crime in Arrests) {
    crime_x = paste(crime, "Disciplinary", sep=".")
    crime_y = paste(crime, "Arrests", sep=".")
    print(combined %>% ggplot(aes(col=Institution.name)) + geom_point(aes_string(x=crime_x, y=crime_y)))
  }
}

compareDiscplinaryArrests(UC_disciplinary, UC_arrests)

Berkeley_NUC_arrests = rbind(Berkeley_arrests, NUC_arrests)

class(Berkeley_arrests)

## [1] "tbl_df"     "tbl"        "data.frame"

class(NUC_arrests)

## [1] "tbl_df"     "tbl"        "data.frame"

Berkeley_summedNUC_arrests = rbind(Berkeley_arrests, sumNonBerkeley(NUC_arrests, FALSE))

massPlot(NUC_arrests, Arrests)

massPlot(NUC_arrests, ArrestsPerStudent)

massPlot(Berkeley_NUC_arrests, Arrests)

massPlot(Berkeley_NUC_arrests, ArrestsPerStudent)

massPlot(Berkeley_summedNUC_arrests, Arrests)

massPlot(Berkeley_summedNUC_arrests, ArrestsPerStudent)

Berkeley_NUC_criminal = rbind(Berkeley_criminal, NUC_criminal)

Berkeley_summedNUC_criminal = rbind(Berkeley_criminal, sumNonBerkeley(NUC_criminal, FALSE))


massPlot(Berkeley_NUC_criminal, Crimes)

massPlot(Berkeley_NUC_criminal, CrimesPerStudent)

massPlot(Berkeley_summedNUC_criminal, Crimes)

massPlot(Berkeley_summedNUC_criminal, CrimesPerStudent)

Berkeley_NUC_disciplinary = rbind(Berkeley_disciplinary, NUC_disciplinary)

Berkeley_summedNUC_disciplinary = rbind(Berkeley_disciplinary, sumNonBerkeley(NUC_disciplinary, FALSE))

massPlot(Berkeley_NUC_disciplinary, Arrests)

massPlot(Berkeley_NUC_disciplinary, ArrestsPerStudent)

massPlot(Berkeley_summedNUC_disciplinary, Arrests)

massPlot(Berkeley_summedNUC_disciplinary, ArrestsPerStudent)